digest

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.30). The workflow requires fetching external RSS feeds from user-defined URLs in memory/feeds.yml and running web/X searches and web fetches, which ingest outsider-authored free text.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.70). The skill instructs the agent to fetch external URLs using curl and WebFetch from memory/feeds.yml, which contains arbitrary user-configurable external URLs, representing a weak/unknown/suspicious operational dependency.